Prognostic Significance Of Platelet Count In Assessing The Severity Of Community Acquired Pneumonia

Abstract

Purpose Platelets are increasingly recognized as inflammatory cell with antimicrobial activity. We hypothesized that abnormal platelet count may be a marker of severity in patients with community-acquired pneumonia (CAP). The objective of this study was to investigate if abnormal platelet count in CAP patients requiring hospitalization was associated with inhospital mortality. Methods It is a retrospective cohort study of all CAP patients hospitalized at Interfaith Medical Center, between May 2009 to August 2010 to analyze the association of platelet count and inhospital mortality. 274 patients were included in the study. Predictor variables were platelet count, leukocyte count, diabetes mellitus, hypertension, chronic obstructive pulmonary disease (COPD) and human immunodeficiency virus (HIV) infection. Abnormal platelet count was defined as platelet count 450,000/L (thrombocytosis). The outcome variable was inhospital mortality. We used SPSS version 19.0 Binomial logistic regression for analysis with forward Likelihood Ratio method. Results Abnormal platelet count was statistically significant with p value of 0.021 with odds ratio of 2.8 for inhospital mortality. PSI score of IV and V was statistically significant as compared to PSI score II with odds ratio of 21 and 56 respectively for mortality but PSI score III was not statistically significant as compared to PSI score II. WBC was not significant for inhospital mortality (0.07). Diabetes mellitus, COPD and HIV infection were not statistically significant for inhospital mortality. Conclusion Abnormal platelet count was associated with high inhospital mortality in patients hospitalized with CAP. So, while evaluating for severity of CAP, platelet count should be used along with PSI score rather than WBC count.